It made Microsoft look bad.

They left out “The Microsoft marketing team called us and complained about our last article. We were so compelled by their plight that we decided to correct ourselves.”

Nvidia wants to be the equal to Intel and AMD. They want to be the 3rd major hardware house.

In ~2009, Intel didn’t renew a contract which allowed Nvidia to produce chipsets for Intel processors, and since then Nvidia has wanted a CPU of their own to keep from getting locked out again.

Nvidia tried to buy Arm when SoftBank was trying to sell, but that got scuttled. They had Tegra in the past which was a phone processor and successful in the Nintendo Switch. They can’t buy Intel because of poison pills in the x86 licensing between AMD and Intel which would kick in.

Fair, the patches don’t have to be accepted. 🙂

Would it be bound by the GPL? Companies love writing shims, and the Linux kernel is pro-business. It is specifically GPLv2 to allow companies to use it in closed source applications. TiVo is the poster child for this.

Anything running in user space isn’t considered a derivative work. The kernel ABI specifically allows for this. A closed source application could run on top of the Linux kernel and not have to be released.

Applications linking to a GPL library, glibc excluded, would have to be released since that would constitute a derivative work.

I’m the PS6 scenario, we would probably get very little usable code. The GPL is old, and companies have had lots of time to work around it.

Companies are just taking BSD code and don’t contribute to it.

There isn’t a lot of evidence of this.

At the end they’re selecting Linux even if there’s licensing risk and they have contribute to code.

This is at odds with the first statement. Companies also aren’t contributing as much code as they should.

Also companies which want to support Linux don’t have to worry that someone would close their code or code they funded with money. It’s not about competition but collaboration.

Yeah, inertia is a thing. It’s why Windows is so dominant. The BSDs were rather competitive with Linux back in the early ‘90s - ‘00s.

This might have been a reason in the ‘90s IBM picked Linux as the Unix successor, but now it’s about inertia and a baseless OS is pretty handy.

It was also never about collaboration. It was always MAD doctrine. Each company had a pack of lawyers ready. The GPL isn’t the most battle tested.

GPL license allowed us also to sell own open-source solutions.

This isn’t the flex you think it is.

Instead of being ready to use solutions…

The BSDs are full operating systems. Batteries are included in the repo.

Linux requires adding lots of other software to make the kernel useful. When people say “It’s GNU/Linux”, this is what they mean. The Linux kernel + the GNU tools make an OS.

…they’re trying to be base for commercial closed-source products and it would be great as contributors could get something from that, but they get nothing.

They are not. They are existing as their own projects. ☺️

Most FOSS devs get nothing. 🤣 GPL, BSD, Apache… It doesn’t matter. The capitalists plunder the commons making money off of other’s hard work.

I understand that BSD see closed source as something cool and way to commercialize software,…

They really don’t. They just want to work on their projects in peace.

…but in today times where a lot of devices have 24/7 access to internet, microphones, cameras and at the same time to sensitive data it’s extremely dangerous. Closed source is used to hide backdoors, acts of surveillance and keeping monopoly on market which obviously stop evolution of software.

Ummm…. That’s not BSD specific. FOSS software gets used for this as well. All those surveillance devices are probably running some sort of Linux.

There are binary firmware blobs and all sorts of stuff. The Linux kernel is GPLv2 specifically to allow this.

The DMCA’s anti-encryption circumvention is used to chill software evolution and lock up code more than anything. BSDs only ask that people don’t GPL their stuff.

Companies play all sorts of games with code, and there isn’t a guarantee that what is in the repo is what people are running. We need reproducible builds to know the software is clean, and without that, software is not trustworthy.

This is bigger than the license on the code. This is about processes and culture.

Please tell me how BSD license can be good solution for operating system.

The code is still open, and the repos will still exist if a fork is created. Sony forks FreeBSD for PS6, and nothing happens to FreeBSD. It still exists, and it still works. The added bonus is not having to deal with Sony, or other people, trying to upstream stuff that doesn’t make sense outside of the PS environment and would have questionable value to others.

There are lots of ways companies get around the GPL, and most are GPL sanctioned.

Starting a company, use the GPL. Starting a project for fun, use whatever because the companies are going to steal it anyway if it’s good.

This is the thing that people overlook. Carrying a patch set is a burden, and companies have to maintain the patch set. Upstream isn’t going to care about the patch set because not their problem, and they will make changes which benefit the open code in their repos.

I’ve seen a company modify a FOSS project to fit their needs and get stuck with a multi-year old version because their changes were incompatible with never versions. They had to scrap the system and start over.

Regardless of license, not contributing upstream creates a problem unless the company has the stomach to support the whole thing on their own.

This is needed. Servers need it, and it would be a nice feature to enable for personal systems. We would need to be able to build our own images with our own keys to really make this worthwhile. Especially with programs in my bin dir I’ve compiled or downloaded.

Do I trust Lennart to not do something asinine to turn this into a shit show? I do not. This would be better if it was someone who has security experience and system design cred.

GPU drivers and DEs lagging behind, mostly.

Something like Fedora which releases newer code quicker will provide a better desktop/laptop experience. It’s the same reason other stable distros, like the EL distros, aren’t the best for desktops/laptops.

Historically, desktop applications would also be versions behind, but Flatpak really helps with this.

At this point, Debian is probably fine as a distro for a few year old computer that won’t be helped by fractional scaling. Pick a DE and install applications from Flathub.

Are someone’s TikToks not trending anymore? 😆

There, there little guy. Why don’t you stand up your own blue sky server?

FreeIPA covers most scenarios. Kerberos, Dynamic DNS/DNS, LDAP.

GPO equivalency would need some config management tool. Ansible is what RH would suggest, but something with an agent would probably be better.

How is this applicable to the comment? Companies never figured out how to charge rent for those.

Devs see home computers as a free resource, and the burden is on the consumer to buy a computer which runs their software.

Nah. Web devs will create even more bloated web pages to keep home computing in business.

For real though, most people don’t need that much computing power, and we reached the plateau 12 years ago. That’s why we’re seeing crypto and AI grifts happen. They recentralize decentralized systems. The elites are striking back.

You know the saying“information wants to be free; information wants to be expensive”? This is the expensive part where people try to horde knowledge by making it inaccessible to everyday people.

Awesome! Previously, Evolution was the only Linux client which supported Exchange, and Evolution is… well…. 😕

That’s going on the list. My heart says I don’t need this, but my brain says I do.

A-series would already be at a disadvantage due to being designed for iPhones and the design parameters that entails compared to the M-series.

The M-series Mac Pro was always for companies which were going to rack them and use them in render farms. Normal people was never its intended market. It was more of an Xserve successor.

Apple would need to design a different CPU for the Mac Pro, and the limited market doesn’t make it feasible. Descending the M-series CPUs from the A-series limits what the designs can do.

There are rumors of a CPU split in the Apple lineup. iPhone, iPad, iMac, Mac Mini, MacBook get the A-series, and MacBook Pro, Mac Studio. Mac Pro get the M-series. That would make sense, and might give them some room to expand the “Pro” procs.

One nit to pick for anyone who reads this later.

/srv is probably a more appropriate location than /mnt. /srv is for local data services are going to serve.

Qualcomm is pretty dumb. Even if this were true, they’d still be leaving Linux support to the community.

The primary ways in which the Mozilla Foundation earns money is through search partnerships, donations and grants.

Yes. It’s the same thing with the Linux kernel and other large FOSS projects. There isn’t a perfect fit for Android, but it would be better than the way ASOP is run now.

As for Red Hat, this comes down to subscriptions or enterprise offerings, neither which really apply to a consumer OS unless you’re willing to pay a subscription fee out of pocket.

Consumer devices ship with proprietary software which is licensed all the time. It could be a library or an entire OS. Consumers are not the target market, like consumers aren’t the target market for RHEL.

The prime example is Windows. It’s licensed to Dell or whomever and ships with the hardware. The license is baked in.

Some people might be willing to pay if the price is reasonable enough. Android has support for major vendors, so using it as a base would be a boon to people doing things like media boxes and signage.

I doubt there will be much to be earned from offering consulting or training, either, unless they make Android exceedingly confusing to use.

It’s the opposite. Make it easy to use. Companies pay for tools which reduces developer time.

The only companies that would pay for Android are OEMs who are already making thin margins, and effectively it’d drive the price of non-iPhones up.

The smaller OEMs would pay for licenses, PS hours, and backend services. They don’t have the expertise or budget.

Samsung? They’re going to keep doing what they’re doing because they have the expertise and budget to fork from upstream. It’s possible they would rally around Android, like companies have rallied around the Linux kernel.

OEMs do this with Linux already, so it would bring Android more inline with the norms.

It could be profitable the way RHEL or the Mozilla Foundation is profitable.

Companies will pay for OS support, and companies will pay for access. Android as a foundation with a company selling OS support and services which could be rebranded would be profitable.

I’m thinking about the wider IoT space here beyond only mobile.